Isotopic evidence for processes of sulfur retention/release in 13 forested catchments spanning a strong pollution gradient (Czech Republic, central Europe)

<jats:p>Sulfur isotope systematics were studied in 13 small catchments in the Czech Republic, similar in topography (V‐shaped valley) and vegetation (Norway spruce). The sites differed in elevation, rainfall, bedrock, soil type and S pollution. Across the sites, δ<jats:sup>34</jats:sup>S values decreased in the order: bulk deposition > runoff > spruce throughfall > C‐horizon soil > A/B‐horizon soil > O‐horizon soil > bedrock (means of 5.5, 4.8, 4.7, 4.6, 4.2, 3.1 and 1.5‰, respectively). Some of the sites had a net export of S, while others accumulated S. Sites exporting S were located in the polluted north where atmospheric S input started to decrease in 1987. Sites retaining S were located in the relatively unpolluted south. Sulfur isotope composition of runoff depended on whether the catchment accumulated or released S. Sites releasing S had runoff δ<jats:sup>34</jats:sup>S values lower than deposition. In contrast, sites retaining S had runoff δ<jats:sup>34</jats:sup>S values higher than deposition. Across the sites, the δ<jats:sup>34</jats:sup>S values of runoff were not correlated with δ<jats:sup>34</jats:sup>S values of bedrock, indicating that the contribution of bedrock to S in runoff was negligible. The δ<jats:sup>34</jats:sup>S values of runoff were strongly positively correlated with the δ<jats:sup>34</jats:sup>S values of soil. Sulfur present in the C‐horizon of soils was mainly derived from atmospheric deposition, not bedrock. Sulfur isotope mass balances were constructed for each catchment, making it possible to quantify the difference between δ<jats:sup>34</jats:sup>S values of the within‐catchment source/sink of S and runoff S. Sulfur isotope mass balances indicated that the sink for the retained S at unpolluted sites and the source of the released S at polluted sites were isotopically fractionated by the same amount relative to runoff S. Inorganic and organic processes were considered as possible causes for this observation. Biological S cycling involves a variety of reactions, some of which fractionate S isotopes. In contrast, adsorption/desorption of inorganic sulfate in soil and weathering of S‐containing minerals do not fractionate S isotopes. Therefore the within‐catchment source/sink of S must be largely a result of biological S cycling. Organic S cycling played an important role over a wide range of atmospheric S inputs from 13 to 130 kg S ha<jats:sup>−1</jats:sup> yr<jats:sup>−1</jats:sup>.</jats:p>